HAL Id: hal-02833143
https://hal.inrae.fr/hal-02833143
Submitted on 7 Jun 2020
HAL
is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire
HAL, estdestinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
management and fire hazard reduction in mediterranean wildland - urban interfaces. Fire Star behaviour model of wildland fire. Experimental values of basic parameters
José Mendes Lopes, Claude Moro, J-Charles Valette, Joao Ventura
To cite this version:
José Mendes Lopes, Claude Moro, J-Charles Valette, Joao Ventura. Fire Star : a decision support system for fuel management and fire hazard reduction in mediterranean wildland - urban interfaces.
Fire Star behaviour model of wildland fire. Experimental values of basic parameters. [Contract]
EVG1-CT-2001-00041/D5-01, 2003. �hal-02833143�
EVG1-CT-2001-00041 D5-01
http://www.eufirestar.org
FIRE STAR:
a decision support system
for fuel management and fire hazard reduction in Mediterranean wildland - urban interfaces
Deliverable D5-01
Fire Star Behaviour Model of Wildland Fire
Experimental Values of Basic Parameters
José MENDES LOPES, Claude MORO,
Jean-Charles VALETTE, João VENTURA,
d5-01.doc
CONTENT LIST
Summary ... 1
Glossary ... 1
List of associated documents ... 1
1 Objectives ... 2
1.1 Flow inside the fuel complex ... 2
1.2 Thermal degradation rate... 2
2 Physical properties of the fuel complex ... 3
2.1 Introduction... 3
2.2 The experimental device... 3
2.2.1 Preparation of the fuel matrix... 3
2.2.2 Main parameters... 4
2.2.3 Velocity measurements... 4
2.2.4 Thermocouples and instrumentation ... 4
2.3 Heat transfer coefficients ... 5
2.3.1 Analysis of the problem... 5
2.3.2 Results ... 5
2.4 Pressure drop ... 6
2.4.1 Pressure measurements ... 6
2.4.2 Temperature measurements ... 6
2.4.3 Calculations ... 6
3 Thermal degradation rate: method... 7
3.1 Procedure... 7
3.1.1 Principle... 7
3.1.2 Preparation of the samples... 7
3.1.3 The measurement itself... 7
3.2 Devices ... 7
4 Thermal degradation rate: results... 8
4.1 Average values ... 8
4.2 Graphs ... 10
SUMMARY The main objectives of the work described in this document are:
- to get information on the flow inside the fuel complex for providing input information to the fire modellers, - to determine the thermal degradation rate of different fuel families.
There are also the main objectives of WP5T1 task.
In section 2, the authors from IST-DEM describe the device they have elaborated for determining the heat transfer coefficients of the different types of wildland fuel beds according to porosity observed or/and measured outside.
They adapt to the device pressures taps along the duct for determining pressure drop per unit of length instead of drag coefficient, which need assumptions about the distribution of the needles.
In sections 3 and 4, INRA’s authors describe the device they are using and the procedure they are following for determining the degradation rate of the different families of wildland fuel particles (leaves, very fine, fine and medium twigs).
The general trends are identical for all the “fine” fuels; coarser particles present a lower degradation rate.
GLOSSARY
porosity of the wildland fuel bed the ratio of the volume of the empty space inside the work module to the total volume of this work module
pressure drop per unit length the ratio of the difference of pressure measured at pressure taps to their distance along the duct; it will be substituted to the drag coefficient in the behaviour model of wildland fire
thermal degradation the decrease of the dimensionless ratio Mt-to-Mi (mass of the sample at the temperature T to the mass of the sample at ambient temperature)
thermal degradation rate this decrease per unit of increasing temperature
LIST OF ASSOCIATED DOCUMENTS None
d5-01.doc 2 1 OBJECTIVES
The main objectives of the work described in this document are:
- to get information on the flow inside the fuel complex for providing input information to the fire modellers,
- to determine the thermal degradation rate of different fuel families.
1.1 FLOW INSIDE THE FUEL COMPLEX
Two important parameters, which are needed to model the flow inside the fuel matrix, are the heat transfer coefficient and the drag coefficient.
The fuel is placed inside a specially designed and built test section, connected to a fan through a heating module.
The controlling parameters are:
- Air flow velocity (used values between 1 ms-1 and 5 ms-1)
- Air temperature (between 293 K and 337 K ) - Fuel bed porosity (varied between 0.92 and 0.94)
The measured properties are:
- Pressure at several points along the working section, in order to get pressure drop values per unit length. Although we have obtained drag coefficients for the initial experiments in this rig, pressure drop per unit length is easier to obtain and does not need assumptions about the needles distribution. This change has been proposed to the modellers and accepted by them.
- Temperature at several points along the working section, from which heat transfer coefficients can be obtained.
There are some properties of the fuel and of the fuel matrix that are provided by our partners in the project (such as fuel density, calorific heat capacity, surface-to-volume ratio, …)
1.2 THERMAL DEGRADATION RATE
The first objective of this operation is to establish the relation between the rate of thermal degradation versus temperature of each type of fuel particle.
The second objective is to compare the results o different fuel particles.
These information may improve the predict ability of the behaviour model of wildland fire in better describing the combustion phase.
2 PHYSICAL PROPERTIES OF THE FUEL COMPLEX
2.1 INTRODUCTION
Surface fire is very important in forest fire ignition and propagation.
The surface fuel bed is a porous medium where a flow of air and gaseous combustion products is established during fire propagation.
The temperatures of the gas and of the fuel bed are not the same, especially in the vicinity of the flame:
therefore, heat transfer occurs.
This exchange of energy plays an important role in the heating (or cooling) of the fuel bed, in the decrease of fuel moisture content and in its pyrolysis.
To model this gaseous flow, values of the coefficient of convection heat transfer and of the drag coefficient of the particles that form the fuel bed matrix are needed.
Figure 2-1: Experimental rig (schematic – not to scale)
Figure 2-2: Work module filled with needles a) with the top cover removed; b) ready to work
2.2 THE EXPERIMENTAL DEVICE
The equipment used to obtain this data will be described in the present section.
The experimental rig used in this work is composed by three modules (Figure 2-1):
The fan module, comprising an electric motor, a fan, a plenum chamber and a contraction bringing the cross section down to 240 mm X 160 mm (width X height).
This equipment, already existing in our lab, was connected to the two following modules, which have been designed and built for the present set of experiments.
The heating module, consisting of a short duct 330 mm long, directly bolted to the contraction exit, contains a set of 15 electrical resistors with an overall power of 5 kW.
The electrical power can be varied by means of an AVAC rheostat.
The work module is a duct with the same cross section (240 mm X 160 mm) and 1.00 m long, made from 1.5-mm thick steel plate.
The top face is removable to allow the duct to be loaded/unloaded with needles.
This top cover is clamped to the main body of the work module.
To avoid the needles to be moved by the air flow, the entrance and exit sections of this module is closed with wire mesh.
The walls of the work module are thermally insulated with a rock wool layer 2-cm thick.
Figure 2-2 shows the work module:
- a) with the top cover removed and - b) ready to work.
The work module is instrumented with five pressure taps and six thermocouples.
2.2.1 Preparation of the fuel matrix
In the laboratory, the needles are kept in cardboard boxes.
Before being loaded into the working section, they are left in the open air.
It can be safely assumed that the needles are in equilibrium with the ambient air.
The needles are weighted to determine the amount necessary to obtain the required matrix porosity.
The top of the working section is then removed and the needles are dropped by gravity.
The needles are uniformly compressed until they all fit inside the working section.
heating elements
honeycomb
thermocouple working section
pine needles heating elements
honeycomb
thermocouple working section
pine needles
preressure
pressure tap
d5-01.doc 4 2.2.2 Main parameters
The main parameters used in the present work are:
- Porosity of the bed, defined as the empty volume inside the work module as a percentage of the total inner volume of this module.
empty total needles
total total
V V V
V V
ε −
= =
- Air flow velocity through the needles matrix: due to the difficulty of measuring inside the bed, flow velocity was measured at the exit of the work module (Uaver) and the velocity inside (Uneedles) was deduced by using continuity equation.
- Air flow temperature.
2.2.3 Velocity measurements
The exit section of the work module is divided in six zones, measurements of velocity are taken in each zone with the thermo-anemometer.
The mean of these six measurements is taken as a representative value (Uaver).
The inlet valve to the fan enclosure controls airflow.
Applying continuity equation between the exit section and a generic cross section inside the work module
( )
exitUaver exit needles Uneedles
A A A
ρ =ρ −
where
- Aneedles is the projected area of the needles in the cross section plane and
- Uneedles is the velocity inside the fuel bed.
For these values of flow velocity, airflow is incompressible, density is constant and after some algebra we get
aver needles
U U
= ε
2.2.4 Thermocouples and instrumentation
The above mentioned six thermocouples are located along the axis of the tunnel.
- the first of them is positioned at the beginning of the working section,
- the last is located 25 mm from the exit, and
- the four remaining ones are equally spaced between those two.
The thermocouples are K type.
The wire has a diameter of 0.25 mm with a spherical joint of 0.70 mm.
These dimensions present the best compromise between mechanical robustness (needed to withstand the loading and removing of the needles) and an acceptable time response.
The wires of the six thermocouples are connected to a Campbell Scientific data logger, model CR10.
This is a programmable unit with a memory capacity of approximately 29 990 values and six connections for voltage measurements in differential mode.
The sampling rate is one measurement per second per thermocouple (i.e. eight measurements per second overall).
The velocity of the air at the exit of the tunnel is measured with a Testo 425 hot wire thermo- anemometer.
The data from the data logger is downloaded into an Excel file.
All the data processing is carried out using the Excel software.
2.3 HEAT TRANSFER COEFFICIENTS 2.3.1 Analysis of the problem
Consider a small volume inside the matrix containing solid and gaseous phases.
An energy balance inside that volume indicates that the amount of thermal energy transferred from the air (gaseous phase) to the needles (solid phase) is equal to the amount of energy that raises the temperature of the needles:
( )
n n p n n n a
n dt
c dT V T
T h
A
=
− ρ
where
- An is the area and Vn is the volume of the needles, - ρn and cp n are the density and the specific heat
capacity of the needles.
The surface to volume ratio of the needles SVRn is known, as well as ρn and cpn.
The temperature of the air Ta can be measured and, hence, the derivative will be known.
The temperature of the needles Tn is very difficult to measure, as well as the derivative (dT/dt)n.
Therefore, it is assumed that (dT/dt)n is equal to (dT/dt)a .
This assumption can be justified because the Biot number of the needles is very low.
The convective heat transfer coefficient h is usually obtained from a Nusselt number.
In the literature, Nu is expressed as a relation between other non-dimensional numbers and some normalised (non-dimensional) characteristics of the geometry and properties of the flow.
Here the results are presented in this standard way.
Special care is needed to obtain the time constant of each thermocouple for each velocity situation.
2.3.2 Results
The value of h is obtained from the energy balance mentioned, and the Nusselt number is then computed.
For Pinus pinaster the following correlation between Nu and Re was obtained for a porosity of 95% and for Re in the range of 50 to 400 taking the linear dimension as 4/SVR of the needles:
6053 .
Re0
1417 . 0 Nu=
Values of h obtained from this correlation are around 30% to 40% of the values found in the literature for packed beds.
This difference was expected due to the much higher capacity of the packed beds (Figure 2-3).
Nu = 0.1417 Re0.6053
R2 = 0.7597
0.0 2.0 4.0 6.0 8.0
0 100 200 300 400 500
Re
Nu
Figure 2-3: Computed values of Nu as a function of Re (Tamb = 22ºC, Thot = 60ºC, matrix porosity = 0.95,
Pinus pinaster needles)
d5-01.doc 6 2.4 PRESSURE DROP
2.4.1 Pressure measurements
The work module was provided with five static pressure taps, connected to a Furnace Control FCO 12 micromanometer, to check linearity in pressure drop along the duct.
After this check, only pressure drop values between the two extreme pressure taps have been used.
Figure 2-4 shows ∆Palong the tunnel as a function of internal velocity for different porosity values.
The adjusted curves are second order polynomials.
2.4.2 Temperature measurements
Some experiments were carried out with airflow temperature in the range 293 to 337 K.
In steady state, the temperature at the exit section was measured with a digital thermometer, and five experiments, separated by 10 K, were performed for each value of porosity.
The set of experiments for each porosity value was carried out with the air inlet valve at a fixed position, corresponding to the middle velocity for the experiments at constant temperature.
It was found that the influence of the temperature on the pressure drop is relatively small for the range of temperatures used.
2.4.3 Calculations
Using the standard definition, the drag coefficient for an average sized needle is
1 2 2
needle D
needles P C F
U A
ρ
=
where
- Fneedle is the force acting on a single needle, - ρ is the air density,
- Uneedles is the flow velocity inside the bed, and - AP is the average projected area, which the needles
present to the flow.
Using average dimensions for the needles (considered as cylinders) and making a simplifying assumption about the needles distribution in the matrix, an expression was obtained for the average drag coefficient of a pine needle in a matrix as a function of flow velocity inside the needles bed and the flow temperature.
(
0.1546 1.3971 0.5087)
0.4913D needles 293
C = − ×U + × T +
This expression is valid for the range of parameters covered by the set of experiments, namely
- Porosity 0.92≤ ≤ε 0.94
- Flow velocity 1.15≤Uneedles
[
m s/]
≤2.48 - Flow temperature 293≤T K[ ]
≤337However, it has been agreed that in future work in this area, results should be expressed as pressure drop per unit length instead of a drag coefficient, as in the first case no assumptions are needed.
It has also been decided not to include the influence of temperature, because in the temperature range of interest, that influence is quite negligible.
20.00 60.00 100.00 140.00
0.50 1.00 1.50 2.00 2.50 3.00
Uneedles [m/s]
∆∆P1,5 [Pa/m]
94% porosity 93%
92%
Figure 2-4: Pressure drop along the work module,
as a function of airflow velocity, for three different values of porosity.
3 THERMAL DEGRADATION RATE: METHOD
3.1 PROCEDURE
It has been described in D6-01 deliverable 3.1.1 Principle
We expose each series of five sample to increasing heat from the ambient to one of the thirteen following levels: from 373 to 973 K, step 50.
We plot the five dimensionless Mt-to-Mi ratios against their temperature level and analyse the shape of the obtained curve.
3.1.2 Preparation of the samples
Before carrying out this measurement, we put the samples in the drying oven (24 hours, 60°C) for working with oven-dried material.
3.1.3 The measurement itself
We constitute 5-g samples of oven-dried material and put them in each series of five cold molybdenum containers.
Then, we put the filled containers in the muffle furnace at ambient temperature and close the front door.
We fix the requested internal temperature and put on the furnace; the temperature increases at 100°K per hour, 1.7 K min-1.
As soon as the internal temperature reaches the studied level (see table here below), we follow the same procedure as for determining ash content:
- to take the hot containers out of the furnace,
- to cover each of the containers with an aluminium sheet for avoiding to loose any particle,
- to place them in the vacuum chamber,
- to wait until the temperature of the containers returns to ambient.
The duration of this last phase depends on the level of the requested temperature.
We determine the “current” mass of each sample on cold material.
We carry out the same operations for each of the thirteen temperature levels.
Effective Temperature in °C
Displayed Temperature at C19 controller in °C
100 105
150 162
200 218
250 273
300 326
350 381
400 436
450 489
500 541
550 593
600 640
650 675
700 700
3.2 DEVICES
Figure 3-1: Electronic balance
Figure 3-2: The furnace and its electronic controller and the five containers
Figure 3-3: Vacuum chamber containing the five containers
d5-01.doc 8 4 THERMAL DEGRADATION RATE: RESULTS
4.1 AVERAGE VALUES
Pinus eldarica needles Avignon Vaucluse
France
Pinus halepensis
needles Avignon Vaucluse
France
Pinus halepensis needles Avignon Vaucluse France collected
in 1999 used in July 2002
Pinus pinaster needles Gargas Vaucluse
France
Pinus pinaster needles
Gargas Vaucluse
France used in July 2000
Pinus pinaster needles
Gargas Vaucluse
France collected in 2002 used in April 2003
Pinus pinea needles Lambert
Var France
Quercus coccifera
leaves Beauchamps
B-d-Rhône France
2001
Quercus coccifera twigs 0-2 Beauchamps
B-d-Rhône France
2001
Quercus coccifera twigs 2-6 Beauchamps
B-d-Rhône France
2001
T en K Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi
373,15 96,91 97,36 - 97,09 96,91 - 96,61 97,07 97,32 98,36
423,15 92,00 91,11 89,69 93,46 92,06 - 92,28 89,38 90,69 92,35
473,15 70,81 71,17 72,59 75,64 73,00 73,82 70,44 70,10 68,75 68,33
523,15 51,05 53,16 54,77 51,43 49,95 51,24 51,63 48,80 50,49 47,30
573,15 40,17 42,72 43,82 40,61 40,63 41,08 37,64 34,48 43,66 37,05
623,15 27,13 31,73 33,97 29,27 29,71 31,27 26,70 16,89 32,01 24,41
673,15 19,43 21,66 23,87 18,98 24,09 22,91 21,30 10,31 20,60 23,29
723,15 14,74 12,84 16,63 10,49 16,49 - 14,61 3,73 15,88 15,12
773,15 9,25 6,51 - 5,71 8,43 - 7,52 3,71 13,53 11,19
823,15 4,74 4,13 - 2,74 4,69 - 4,53 3,37 7,29 8,82
873,15 3,97 3,71 - 2,09 2,54 - 4,19 2,90 4,56 5,46
923,15 3,33 3,07 - 2,07 2,38 - 4,05 2,74 3,70 3,70
973,15 3,48 3,44 - 2,23 2,30 - 3,93 2,83 3,64 3,49
Quercus ilex leaves Trou du rat
Lubéron Vaucluse
France summer 2002
Quercus ilex twigs 0-2 Trou du rat
Lubéron Vaucluse
France summer 2002
Quercus ilex twigs 2-6 Trou du rat
Lubéron Vaucluse
France summer 2002
Quercus ilex twigs 6-25 Trou du rat Lubéron Vaucluse
France summer 2002
Quercus suber bark (Saltus)
Pinus halepensis bark (Saltus)
Pinus halepensis écailles of the
cone (Saltus)
Pinus halepensis twigs 2-6 (saltus)
(Saltus)
T en K Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi Mt/Mi
373,15 96,65 - 97,01 97,74 98,66 97,19 97,69 97,60
423,15 90,44 - 91,17 92,82 96,26 90,02 93,34 90,57
473,15 67,29 69,58 67,57 68,08 84,15 69,26 68,91 63,00
523,15 48,87 52,23 48,98 45,45 73,08 57,67 49,48 40,31
573,15 38,67 43,70 41,52 38,17 50,13 46,11 40,60 32,19
623,15 25,32 35,16 32,44 33,11 35,16 35,38 28,67 24,77
673,15 21,02 - 25,85 26,84 24,55 28,15 19,57 18,61
723,15 14,25 - 21,47 21,42 14,71 22,11 12,67 9,01
773,15 4,36 - 16,61 16,59 8,26 11,94 4,69 7,89
823,15 4,23 - 12,00 14,94 5,16 10,13 1,80 5,65
873,15 3,63 - 9,21 12,30 1,56 4,17 0,48 2,90
923,15 2,89 - 7,20 10,34 1,33 3,22 0,53 2,59
973,15 2,92 - 6,33 6,00 1,63 2,90 0,54 2,39
d5-01.doc 10 4.2 GRAPHS
Thermal degradation rate
0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 90,00 100,00
300 400 500 600 700 800 900 1000
Temperature (K) Mt/Mi
Pinus eldarica needles Pinus halepensis needles Pinus pinaster needles Pinus pinaster needles Pinus pinea needles
Figure 4-1: Thermal degradation rate for pine needles
Thermal degradation rate
0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 90,00 100,00
300 400 500 600 700 800 900 1000
Temperature (K) Mt/Mi
Quercus coccifera leaves Quercus coccifera twigs 0-2 Quercus coccifera twigs 2-6
Figure 4-2: Thermal degradation rate of particles families of Quercus coccifera
Thermal degradation rate
0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 90,00 100,00
300 400 500 600 700 800 900 1000
Temperature (K) Mt/Mi
Quercus ilex leaves Quercus ilex twigs 0-2 Quercus ilex twigs 2-6 Quercus ilex twigs 6-25
Figure 4-3: Thermal degradation rate of particles families of Quercus ilex
Thermal degradation rate
0,00 10,00 20,00 30,00 40,00 50,00 60,00 70,00 80,00 90,00 100,00
300 400 500 600 700 800 900 1000
Temperature (K) Mt/Mi
Quercus suber bark Pinus halepensis bark Pinus halepensis cone Pinus halepensis twigs 2-6
Figure 4-4: Thermal degradation rate of miscellaneous fuel particles families
